144 GLACIAL FORMATIONS OF ERIE AND OHIO BASINS. 



prior to the gravel deposition seems to indicate that it stood above the old 

 stream for a considerable period before the gravel deposition occurred. But 

 it does not throw light upon the depth to which trenching had reached. 



Within the trench cut in the old gradation plain there are a few rock 

 shelves which seem sufficiently near a definite horizon to be correlated. 

 They stand about 75 to 100 feet above the river, and vary in width from 

 one-fourth of a mile to narrow strips but a few feet wide. Whether they 

 are merely an incident in the cutting down of the valley or signify the ter- 

 mination of an epoch of degradation which was followed by a notable halt 

 and possibly a refilling, has not been determined. 



There remain two other fluvial plains to be considered. One is the 

 rock floor beneath the present stream and the other is the gravel filling 

 which took place in connection with the Wisconsin glaciation. The rock 

 floor is usually but 20 to 30 feet below the stream bed, though in a few 

 places it appears to reach 50 feet. The excavation down to the rock floor 

 seems to have preceded the Wisconsin stage of glaciation, for the rock floor 

 is found to be as low under undisturbed portions of the Wisconsin gravel 

 as in the trench which the river has cut in that gravel. The filling connected 

 with the Wisconsin glaciation extends to a height of 60 to 80 feet above the 

 present stream, and about 100 feet above the rock floor. 



In the table below the relation and altitudes of the several fluvial plains 

 are set forth. The altitudes are largely, barometric, but as a base for cal- 

 culation the Allegheny Valley Railway has furnished a series of levels 

 extending the whole length of the Lower Allegheny. The altitudes of the 

 stream, the sui-face of the Wisconsin terrace, and the rock floor below the 

 stream were estimated with a fair degree of accuracy from these railway 

 levels, but for the gradation plain, high rock shelves, and the upper limit 

 of gravel on the gradation plain, the barometer was called into use. In the 

 vicinity of Pittsburg, however, Jillson has made a series of measurements 

 of the gradation plain and upper hmit of gravel with a Locke level. 



